PROM -

[sashura]

Prelabor Rupture of the Membranes at Term (Correspondence) Ennis, Michael C.; Geary, Michael P.; Morrison, John J.; Ecker, Jeffrey L.; Kilpatrick, Sarah J.; Hannah, Mary E.; Duff, Patrick.
The New England Journal of Medicine , Oct 10 , 1996 , Volume: 335 , Number: 15,
Page: 1156-1159

Letter 001

To the Editor: The study by Hannah et al. (April 18 issue) (Ref. 1) exemplifies the best of evidence-based medicine: it is a multicenter, randomized, prospective trial with a large patient population and blinded outcome assessment. Despite the trial's methodologic and statistical strengths, the fact that the authors did not control for parity, cervical ripeness, digital vaginal examinations, or the use of oxytocin in their expectant-management group raises questions regarding the validity of their conclusions. The study fails to isolate and address the most vexing problem related to prelabor rupture of the membranes: how to care for a nulliparous woman with an unfavorable, or unripe, cervix (defined in this study as a cervix that is dilated less than 3 cm and is less than 80 percent effaced).

Prospective studies comparing expectant management and the induction of labor in low-risk women with prelabor rupture of the membranes whose cervixes were unfavorable note a higher rate of cesarean section and perinatal infection in patients treated by induction with oxytocin. (Ref. 2,3) Conversely, a woman with prelabor rupture of the membranes whose cervix is favorable will probably do well with induction of labor. Other studies suggest that nulliparous women in particular fare less well if labor is induced because of prelabor rupture of the membranes. (Ref. 4) Several factors may have contributed to the apparent lack of differences in cesarean-section rates in the study by Hannah et al.
Multiparas (40 percent of the women) were pooled with nulliparas (60 percent of the women). Women with favorable cervixes and those whose cervixes were unfavorable were both included. On the basis of a digital vaginal examination or vaginal examination with a speculum, cervical status was considered ripe in
6 to 15 percent of patients, unripe in 29 to 54 percent and, surprisingly, unknown in 33 to 65 percent. Previous studies have used a Bishop score of less/equal 4 as the criterion for an unfavorable cervix. Hannah et al. used liberal criteria (dilated greater/equal 3 cm and greater/equal 80 percent
effaced) to define a ripe cervix.

Another major issue for clinicians dealing with prelabor rupture of the membranes is infection. Earlier studies that demonstrated a lower rate of perinatal infectious morbidity in the expectant-management group defined expectant management as observing a patient with prelabor rupture of the membranes until the baby was delivered or complications arose. (Ref. 2,3) Hannah et al. allowed expectant management to continue for only four days. A striking 49.9 percent of their expectant-management group eventually received oxytocin. This approach is not truly expectant; rather, it more closely approximates a delayed induction. Such an approach is associated with more febrile morbidity than either early induction or true expectant management.
(Ref. 5)

Finally, two of the strongest associations between infection and prelabor rupture of the membranes are the time from the first digital vaginal examination until delivery and the total number of such examinations. (Ref. 6) Frequent digital vaginal examinations are usually necessary during the induction of labor. A cornerstone of previous research on expectant management involved minimizing the number of such examinations. Dr. Hannah and her associates stated that ``digital vaginal examinations were avoided.'' However, their expectant-management group underwent significantly more digital vaginal examinations than their induction-with-oxytocin group.

By pooling data on patients with these confounding factors Hannah et al. may have achieved statistical power while quite possibly diluting and obscuring findings relevant to managing the most challenging problems associated with prelabor rupture of the membranes at term.

Michael C. Ennis, M.D.University of Massachusetts Medical CenterWorcester, MA
01655-0309

Letter 002

To the Editor: There are two points we would like to address with respect to the paper by Hannah et al.

First, for the women in the induction-with-prostaglandin group, 1 or 2 mg of prostaglandin E(sub 2) gel was inserted into the posterior vaginal fornix, and this application was repeated six hours later if labor had not started. Some women thus received a total of 2 mg, and others received 4 mg. In view of the fact that the cervical status was unknown in 33 percent to 35.8 percent of the women, this may have confounded the results. Hence, the findings that showed that the induction-with-oxytocin group had shorter labors, shorter intervals between rupture of the membranes and delivery, and shorter intervals between admission and delivery than the induction-with-prostaglandin group are open to question and do in fact contradict previously published data. (Ref. 1)

Second, the results of this important study indicate that expectant management is associated with an increased frequency of maternal and neonatal infection, suggesting that obstetricians have a responsibility to induce labor immediately after the rupture of membranes at term, rather than wait as long as four days.
(Ref. 2) However, expectant management for up to 24 or 48 hours has been reported to confer significant advantages without producing additional hazard, since it leads to reduced cesarean-section rates (Ref. 3,4) and minimal (Ref.
3) or no (Ref. 4) clinical morbidity. It is our opinion that such an important and expensive trial should have been based on currently available literature and hence directed at finding solutions to existing management problems (i.e., induction vs. expectant management for up to 24 to 48 hours) rather than directed solely at the creation of novel alternatives (i.e., expectant management for up to four days). It is unfortunate that the authors did not include in the trial a more realistic and clinically acceptable interval of 48 hours of expectant management, for example, as well as that of four days. The results of such a comparison might have yielded very different findings.

Michael P. Geary, M.D.John J. Morrison, M.D.University College LondonLondon WC1E 6AU, United Kingdom

Letter 003

To the Editor: Although Hannah et al. report a higher rate of infectious morbidity among women in whom prelabor rupture of the membranes was managed expectantly, this finding may well be attributable to the protocols and definitions used in the study. Specifically, we are concerned that the authors'
definition of chorioamnionitis was too liberal, leading to the inclusion of many women with fever but no other signs of intrauterine infection. We also note that a temperature cutoff of 37.5 degreesC was used to define antepartum and intrapartum fevers, a level lower than that used in most standard definitions. (Ref. 1) We are also troubled that at least 40 percent of the women in the expectant-management group underwent a digital vaginal examination, a known risk factor for subsequent infection. (Ref. 2) Even with these definitions and risks, the study demonstrated differences in the rates of febrile morbidity only when the active- and expectant-management groups randomly assigned to receive oxytocin were compared, and no increase was found in neonatal morbidity of any kind in any group.

Recognizing the similarity of outcomes in their study, the authors appropriately conclude that both ``induc-tion... and expectant management are... reasonable options.'' In contrast, in the editorial accompanying this study, Dr. Duff states that ``expectant management, followed by delayed induction of labor... is a practice that should be abandoned.'' (Ref. 3) It seems to us that such a conclusion promotes a standard that is not supported by the available evidence and unnecessarily limits clinicians' and patients'
choices. Importantly, nearly 80 percent of patients in the expectant-management group began labor on their own. Although as a whole the patients preferred active rather than expectant management, individual women may prefer to wait:
certainly, the experience of an induced labor differs from that of spontaneously occurring contractions, especially if expectant management is conducted at home. We suggest that both expectant and active management of prelabor rupture of the membranes at term are acceptable plans in appropriately counseled patients.

Jeffrey L. Ecker, M.D.Sarah J. Kilpatrick, M.D., Ph.D.University of California, San FranciscoSan Francisco, CA 94143-0132

The authors reply:

Letter 004

To the Editor: My colleagues and I do not share Dr. Ennis's concern that the results of our study are not valid for nulliparous women with unfavorable cervixes. There were no significant differences between groups in parity, the numbers of previous vaginal examinations, or the degree of cervical ripeness at entry into the study. All treatments or tests after randomization were a result of the protocols for active or expectant management. A total of 3000 women were nulliparous, and in 64.2 percent of these, the cervix was dilated less than 3 cm and was less than 80 percent effaced (on the basis of a digital vaginal examination, vaginal examination with a speculum, or both). The rates of cesarean section among nulliparous women with unfavorable cervixes in the induction-with-oxytocin, induction-with-prostaglandin, expectant-management (oxytocin), and expectant-management (prostaglandin) groups were 14.8 percent
(71 of 480 women), 14.1 percent (68 of 483), 15.0 percent (70 of 467), and 14.9 percent (74 of 497), respectively. Other studies have not consistently found higher rates of cesarean section after induction with oxytocin than after expectant management in nulliparous women with unfavorable cervixes, and several of the reports cited by Dr. Ennis are methodologically of very poor quality.

The length of expectant management was not limited to a specific number of hours, since we had no evidence that active management was necessary at a particular point in time. We offered induction of labor to women in the expectant-management groups after four days because it was our experience that women were reluctant to continue expectant management beyond this.

Vaginal examinations were least frequent in the induction-with-oxytocin group.
Perhaps this was because active labor and the time to active labor were shortest in this group. Our finding of a shorter labor and a shorter time to active labor with induction with oxytocin as compared with prostaglandin is consistent with previous reports. (Ref. 1,2) We allowed clinicians to choose the dose (1 mg or 2 mg) of prostaglandin gel they thought was most appropriate.
Factors that may have influenced their choice were usual practice, the state of the cervix, and the woman's parity.

We agree completely with Drs. Ecker and Kilpatrick that after women have been informed of the benefits and risks of the different options, both expectant and active management should be offered to women at term with prelabor rupture of the membranes.

Mary E. Hannah, M.D.C.M.University of Toronto Perinatal Clinical Epidemiology UnitToronto, ON M5G 1N8, Canada

Letter 005

To the Editor: In an article published in 1984, (Ref. 1) my colleagues and I found that women with prelabor rupture of the membranes at term and unfavorable cervixes who were cared for expectantly as inpatients until spontaneous labor or infection ensued had a lower frequency of cesarean delivery and chorioamnionitis than women treated by immediate induction of labor.
Subsequently, however, Wagner et al. (Ref. 2) noted a significantly higher frequency of suspected neonatal sepsis when labor was induced in similar patients 24 hours after rupture of the membranes rather than within 6 hours afterward. In another related study, Guise et al. (Ref. 3) reported a higher frequency of maternal infection and suspected neonatal sepsis in women who were observed for 24 hours and then underwent induction. In a more recent report, Shalev and coworkers (Ref. 4) conducted a prospective, but nonrandomized study of women with unfavorable cervixes who were not in labor within six hours after rupture of the membranes. Patients were assigned to induction at the end of either 12 or 72 hours. Although the two groups did not differ in terms of the rates of cesarean delivery or maternal and neonatal infection, patients who underwent induction after 72 hours had a significantly longer and more expensive hospitalization.

In the study by Hannah et al., (Ref. 5) patients in the expectant-management
(oxytocin) group had a significantly higher frequency of infection than those in the induction-with-oxytocin group. Their infants also were significantly more likely to receive antibiotics for suspected sepsis than infants in the induction-with-oxytocin group. Moreover, four perinatal deaths occurred in the two expectant-management groups, as compared with none in the induction groups.
Although this difference in perinatal mortality was not statistically significant, a much larger sample size would be necessary to exclude definitively the possibility of a type II statistical error. Finally, Hannah and coworkers did not examine the cost of alternative plans of management, an extremely important issue in today's economic climate.

I agree that patients should be informed of all reasonable treatment options.
However, in view of the possible risks of delayed induction outlined above, the unproven safety of outpatient expectant management, the probable increased cost of inpatient expectant management, and the clear preference of patients in the study by Hannah et al. for the immediate induction of labor, I believe a more proactive approach to this difficult clinical problem is appropriate. From my perspective at least, an initial period of expectant management followed by induction of labor at some arbitrary interval seems to involve more risk than benefit to the mother and her baby.

Patrick Duff, M.D.University of Florida College of MedicineGainesville, FL
32610-0294

Bibliography:

Reference 001
Hannah ME, Ohlsson A, Farine D, et al. Induction of labor compared with expectant management for prelabor rupture of the membranes at term. N Engl J Med 1996;334:1005-10.Reference 002 Duff P, Huff RW, Gibbs RS. Management of premature rupture of membranes and unfavorable cervix in term pregnancy. Obstet Gynecol 1984;63:697-702.Reference
003
Morales WJ, Lazar AJ. Expectant management of rupture of membranes at term.
South Med J 1986;79:955-8.Reference 004
Sperling LS, Schantz AL, Wahlin A, et al. Management of prelabor rupture of membranes at term: a randomized study. Acta Obstet Gynecol Scand 1993;72:627-32.Reference 005 Wagner MV, Chin VP, Peters CJ, Drexler B, Newman LA. A comparison of early and delayed induction of labor with spontaneous rupture of membranes at term.
Obstet Gynecol 1989;74:93-7.Reference 006 Schutte M, Treffers PE, Kloosterman GJ, Soepatmi S. Management of premature rupture of membranes: the risk of vaginal examination to the infant. Am J Obstet Gynecol 1983;146:395-400.Reference 007 Goeschen K. Premature rupture of the membranes near term: induction of labor with endocervical prostaglandin E2 gel or intravenous oxytocin. Am J Perinatol 1989;6:181-4.Reference 008 Duff P. Premature rupture of the membranes at term. N Engl J Med 1996;334:1053-4.Reference 009 Kappy KA, Cetrulo CL, Knuppel RA, et al. Premature rupture of the membranes at
term: a comparison of induced and spontaneous labors. J Reprod Med 1982;27:29-33.Reference 010 Conway DI, Prendiville WJ, Morris A, Speller DC, Stirrat GM. Management of spontaneous rupture of the membranes in the absence of labor in primigravid women at term. Am J Obstet Gynecol 1984;150:947-51.Reference 011 Blanco JD. Intra-amniotic infections. In: Gleicher N, ed. Principles and practice of medical therapy in pregnancy. 2nd ed. Norwalk, Conn.: Appleton & Lange, 1992:714.Reference 012 Schutte MF, Treffers PE, Kloosterman GJ, Soepatmi S. Management of premature rupture of membranes: the risk of vaginal examination to the infant. Am J Obstet Gynecol 1983;146:395-400.Reference 013 Duff P. Premature rupture of the membranes at term. N Engl J Med 1996;334:1053-4.Reference 014 McQueen D, Neilson JP, Whittle MJ. Pre-labour rupture of membranes with an unripe cervix: a random trial of management. J Obstet Gynecol 1990;10:495-8.Reference 015 Chua S, Arulkumaran S, Kurup A, Anandakumar C, Tay D, Ratnam SS. Does prostaglandin confer significant advantage over oxytocin infusion for nulliparas with pre-labor rupture of membranes at term? Obstet Gynecol 1991;77:664-7.Reference 016 Duff P, Huff RW, Gibbs RS. Management of premature rupture of membranes and unfavorable cervix in term pregnancy. Obstet Gynecol 1984;63:697-702.Reference
017
Wagner MV, Chin VP, Peters CJ, Drexler B, Newman LA. A comparison of early and delayed induction of labor with spontaneous rupture of membranes at term.
Obstet Gynecol 1989;74:93-7.Reference 018 Guise JM, Duff P, Christian JS. Management of term patients with premature rupture of membranes and an unfavorable cervix. Am J Perinatol 1992;9:56-60.Reference 019 Shalev E, Peleg D, Eliyahu S, Nahum Z. Comparison of 12- and 72-hour expectant management of premature rupture of membranes in term pregnancies. Obstet Gynecol 1995;85:766-8.Reference 020 Hannah ME, Ohlsson A, Farine D, et al. Induction of labor compared with expectant management for prelabor rupture of the membranes at term. N Engl J Med 1996;334:1005-10.


исходная статья:
http://content.nejm.org/cgi/content/abstract/334/16/1005